Overweight,Obesity and Endometrial Cancer Risk: Results from a Systematic Review and Meta-Analysis

Abstract

Aim

Findings from recent studies suggest that obesity may be associated with an increased risk of endometrial cancer, but several earlier studies were less conclusive. Here we strive to estimate this relationship in a meta-analysis of published data.

Methods

We searched Pubmed and Embase for studies on body mass index and the risk of endometrial cancer, published from 1989 to 2011. Data were independently extracted and analyzed using random or fixed effects meta-analysis depending on the degree of heterogeneity.

Results

Seven cohort studies and 11 case-control studies were included in the meta-analysis. Overall, the conditions of excess body weight ([EBW] defined as body mass index [BMI] ≥25 kg/m²), obesity (BMI ≥30 kg/m²) and overweight (25< BMI <30 kg/m²) were associated with an increased risk of endometrial cancer (relative risk [RR] for EBW=1.62, 95% confidence interval [CI], 1.39-1.89; for obesity RR=2.54, 95% CI, 2.11-3.06; for overweight RR=1.32, 95% CI, 1.16-1.50). Subgroup analyses showed that the positive associations were independent of study design, geographic locations, self-reported BMI, alcohol use, smoking habit, history of diabetes, hormone therapy, age at menarche, age at menopause, parity, and age at first full term pregnancy. However, there was no statistically significant association between EBW and endometrial cancer risk for measured BMI (for EBW RR=1.29, 95% CI, 0.66-2.53).

Conclusions

The findings from this meta-analysis strongly support that the conditions of EBW, overweight, and obesity are all associated with an increased risk of endometrial cancer. Also, the strength of the association increases with increasing BMI.

Keywords

Endometrial cancer Excess body weight Obesity Overweight Meta-analysis Relative risk

Introduction

The International Agency for Research on Cancer recently estimated that endometrial carcinoma is the most common gynecological cancer in the developed world (1). Endometrial cancer is the third most common cancer in women in Western countries, accounting for 6%-9% of all cancer types in women (2). It was estimated that in 2006 a total of 40,200 women in the US were affected by endometrial cancer (3), and that women in the US have a 2.6% lifetime risk of being diagnosed with this disease (4). Endometrial cancer is often diagnosed at an early stage, and has a high cure rate (5). Thus, enhancing the survivors’ quality of life is of high priority. Therefore, early intervention on modifiable risk factors for endometrial cancer is very important.

Obesity is an increasing problem in the US. About 65% of the population in the US is overweight and 30% of the population is obese, according to their body mass index (BMI) (6). Obesity affects more than 25% of adult women in the US and continues to increase in prevalence. Numerous epidemiological studies have demonstrated that obesity is a major risk factor for endometrial cancer (7). Although an average woman has a 3% lifetime risk of developing endometrial cancer, this risk raises to 9%-10% for obese women (8). However, the recently published studies on the association between excess body weight (EBW) and the risk of endometrial cancer have no consistent conclusion.

As to evaluate the association between EBW and the risk of endometrial cancer, we carried out a systematic review and meta-analysis of published cohort studies and case-control studies.

Materials and Methods

Search Strategy

We identified eligible studies (published up to the 15^th of February 2011) by performing a comprehensive search in Pubmed and Embase. Research papers were selected using the following sets of keywords: “body mass index” or “BMI” or “obesity” or “adiposis” or “excess body weight”; “endometrial cancer” or “endometrial carcinoma” or “endometrial adenocarcinoma” or “endometrial tumor” or “endometrial neoplasm” or “EC”; “risk” or “incidence” or “mortality”. The search was performed on studies conducted on human subjects, with restriction on English language papers. We reviewed all abstracts obtained from our search for relevance, we manually reviewed bibliographies and review articles for additional citations, and obtained the full text of all potentially relevant articles. We did not consider abstracts or unpublished reports.

Inclusion and Exclusion Criteria

We evaluated all of the studies retrieved according to our selection criteria. The following criteria were used to include published studies: (i) the study had to be a cohort or case-control study examining exposure to EBW or high BMI and the incidence or mortality rates of endometrial cancer; (ii) the study had to contain an explicit definition of BMI, as well as of obesity and overweight conditions, with raw data for each group. The major reasons for exclusion of studies were: (i) no raw data available; (ii) duplicate studies; (iii) the data reported were not usable.

Data Extraction

Two investigators independently extracted the data, discussed disagreements, and reached consensus on all issues. From each study considered the following information was extracted using a standardized data-collection protocol: the first author's name, year of publication, country of origin, study design (cohort versus case-control), period of diagnosis, sample size, exposure measurement method (self-reported versus measured BMI versus diagnosis of obesity), exposure definitions (BMI definitions of overweight or obese), and variables for which the analysis was adjusted. Seven cohort studies and 11 case-control studies were included in the meta-analysis. All studies have a high quality evaluation, in each research, such as taking measures to control for confounding factors, describing the results of the lost object, having enough time to review. Some of the studies were published in high-impact journals (such as the Journal of the National Cancer Institute, Cancer Research, the International Journal of Cancer, and the International Journal of Epidemiology). For studies including subjects of different ethnic groups, when possible, data were separately extracted for each ethnic group. We did not define any minimum number of patients for a study to be included in our meta-analysis.

Exposure Definition

We defined body mass groups using the following BMI categories: normal weight (BMI ≥18.5 kg/m² and <25 kg/m²), overweight (BMI ≥25 kg/m² and <30 kg/m²), and obese (BMI ≥30 kg/m²). Combined overweight and obesity may be expressed as EBW (BMI ≥25 kg/m²). These groupings represented the divisions or quartiles most frequently reported in the articles included, and may somewhat differ from the BMI categories defined in other studies (9). For each study, we selected the BMI-cancer association that most closely approximated each of these categories and the date of BMI determination that was most commonly reported.

Statistical Analysis

The strength of the association between the EBW and the risk of endometrial cancer was measured by risk ratios (RRs) with 95% confidence intervals (CIs). Pooled estimates of the average RR were obtained by calculating a weighted average of the RRs from each study. Either the random-effects or the fixed-effects model were used for the meta-analysis. When the I² statistic was >50%, a random-effects model was chosen, otherwise a fixed-effects model was used. Heterogeneity among studies was assessed using Q and I² statistics, which test the total variation across studies that was attributable to heterogeneity rather than to chance (10). Funnel plots and test statistics were used to determine the existence of potential publication bias. All analyses were performed with Review Manager 5.0 software (The Cochrane Collaboration, Oxford, United Kingdom, 2008).

Results

Study Characteristics

There were 395 published papers relevant to the search terms. After screening the title, 192 papers were excluded. Thus, 203 potentially eligible studies were reviewed and, after this step, additional 185 studies were excluded, leading to a final number of 18 papers included. Of these, 7 cohort studies and 11 case-control studies were included in the pooled analyses (9, 11 –27). The baseline characteristics of the selected studies included in the meta-analysis are summarized in Table I. Figure 1 describes the results of the systematic literature search papers.

Table I

Characteristics of Cohort and Case-Control Studies Included in this Meta-Analysis

Authors	Year	Country	Design	Period of diagnosis	Case ascertainment	No of cases	Exposure	BMI reference	BMI overweight	BMI obese	BMI overweight and obese	Adjustments^*
Lukanova A (11)	2006	Sweden	Cohort	1985–2003	Cancer registry	N=118 total N=35362	Measured BMI at enrollment in the study	18.5–24.9	25–29.9	≥30	≥25	1,2
Reeves CK (12)	2010	UK	Cohort	1996–2001	Cancer registry	N=2657 total N=1222630	Self-reported BMI at enrollment in the study	22.5–24.9	25–29.5	≥30	≥25	1, 2, 3, 4, 5, 6, 7, 8, 13
Chang SC (13)	2007	California	Cohort	1995–2000	Cancer registry	N=677 total N=103882	Measured BMI at enrollment in the study	<25	25–29.9	≥30	≥25	1, 2, 7, 8, 9, 11, 12, 13, 14, 15
Canchola AJ (14)	2010	California	Cohort	1995–2006	Cancer registry	N=395 total N=28418	Self-reported BMI at enrollment in the study	<25	25–29	≥30	≥25	1, 7, 11, 12, 14, 16, 17
Jain MG (15)	2000	Canada	Cohort	1980–1993	Committee of physicians	N=221 total N=3918	Measured BMI at enrollment in the study	<25			≥25	1, 2, 10, 11, 12, 13, 14, 18
Conroy MB (16)	2009	US	Cohort	1992–2007	Cancer registry	N=264 total N=32642	Self-reported BMI at enrollment in the study	22.5–24.9	25–29.9	≥30	≥25	1, 2, 6, 7, 8, 13, 14
Schouten LJ (17)	2004	Netherlands	Cohort	1986–1995	Cancer registry	N=226 total N=62573	Self-reported BMI at enrollment in the study	20–24.9	25–29 .9	≥30	≥25	1, 2, 7, 8, 11, 12, 14
Xu WH (18)	2006	China	Case-control	1997–2001	Cancer registry	Cases 832 controls 846	Self-reported BMI at enrollment in the study	21.04–25.69			≥25.69	1, 2, 8, 12, 17, 19
Lu L (19)	2011	Yale	Case-control	2004–2009	Physician consent	Cases 668 controls 674	Measured BMI at enrollment in the study	<25	25–30	≥30	≥25	1, 2, 9, 10, 12, 13, 17, 19
Hosono S (20)	2011	Japan	Case-control	2001–2005	Cancer center hospital	Cases 222 controls 2162	Self-reported BMI at enrollment in the study	<25			≥25	1, 2, 6, 7, 8, 11, 12, 13, 14, 15
Folsom AR (21)	1989	Iowa	Case-control	1980–1986	Cancer registry	Cases 63 controls 1274	Self-reported BMI at enrollment in the study	<24.31	24.31–28.33	28.33	≥24.31	1
Charneco E (22)	2010	Puerto Rico	Case-control	2004–2007	Cancer registry	Cases 74 controls 88	Self-reported BMI at enrollment in the study	<24.9	25–29.9	≥30	≥25	1, 4, 10, 12, 15, 16
Dal Maso L (23)	2004	Italy	Case-control	1999–2002	Cancer registry	Cases 87 controls 132	Self-reported BMI at enrollment in the study	<25	25–29	≥30	≥25	1, 10
John EM (24)	2010	San Francisco	Case-control	1996–1999	Cancer registry	Cases 472 controls 443	Self-reported BMI at enrollment in the study	<25	25–29.9	≥30	≥25	1, 9
Fortuny J (25)	2009	US	Case-control	2001–2005	Cancer registry	Cases 469 controls 467	Self-reported BMI before diagnosis	<25	25–30	≥30	≥25	1
Okamura C (26)	2006	Japan	Case-control	1998–2000	Cancer registry	Cases 155 controls 96	Self-reported BMI at enrollment in the study	20.04–23.92			≥23.92	1
Trentham-Dietz A (27)	2006	US	Case-control	1991–1994	Surgery diagnosis	Cases 740 controls 2342	Self-reported BMI at enrollment in the study	22.6–25.4	25.5–29.2	≥29.2	≥25.5	1, 2, 7, 8, 11, 13, 14, 15
Potischman N (28)	1996	US	Case-control	1987–1990	Cancer registry	Cases 276 controls 316	Self-reported BMI at enrollment in the study	23–26	26–30	≥30	≥26	1, 9, 10

1. age; 2. smoking; 3. geographical region; 4. socioeconomic status; 5. reproductive history; 6. alcohol intake; 7. physical activity; 8. age at menopause; 9. ethnicity; 10. education; 11. age at menarche; 1 2. oral contraceptive use; 13. duration and type of hormone therapy; 14. parity; 15. diabetes; 16. hypertension; 17. age at first full term pregnancy; 18. total energy; 19. family history of cancer.

Fig. 1

Results of the systematic literature search: number of studies and their inclusion or exclusion.

Meta-Analysis

The combined results of all the studies considered showed that there was a statistically significant association between EBW, overweight, or obesity and endometrial cancer risk (for EBW RR=1.62, 95% CI, 1.39–1.89; for obesity RR=2.54, 95% CI, 2.11–3.06; for overweight RR=1.32, 95% CI, 1.16–1.50). The strength of the association increased with increasing BMI (Tab. II, Figs. 2–4).

Table II

Statistical Results of Cohort and Case-Control Studies Included in the Meta-Analysis by Revman 5.0

Groups	No of studies	RR (95% CI)	Statistical method	P
All studies
EBW	18	1.62 (1.39–1.89)	Random	<0.00001
Obese	14	2.54 (2.11–3.06)	Random	<0.00001
Overweight	14	1.32 (1.16–1.50)	Random	<0.0001
Cohort
EBW	7	1.62 (1.14–2.30)	Random	0.008
Obese	6	3.11 (2.62–3.69)	Random	<0.00001
Overweight	6	1.47 (1.22–1.78)	Random	<0.0001
Case-control
EBW	11	1.60 (1.39–1.84)	Random	<0.00001
Obese	8	2.06 (1.61–2.63)	Random	<0.00001
Overweight	8	1.18 (0.96–1.45)	Random	0.12^#
Asians
EBW	5	1.68 (1.35–2.09)	Random	<0.00001
Obese	2	2.65 (1.34–5.24)	Random	0.005
Overweight	2	1.80 (0.89–3.67)	Random	0.1
Non-Asians
EBW	13	1.59 (1.29–1.95)	Random	<0.0001
Obese	12	2.52 (2.04–3.12)	Random	<0.00001
Overweight	12	1.29 (1.12–1.48)	Random	0.0004
BMI ascertainment
Self-reported
EBW	14	1.70 (1.49–1.94)	Random	<0.00001
Obese	11	2.46 (1.96–3.08)	Random	<0.00001
Overweight	11	1.29 (1.10–1.52)	Random	0.002
Measured
EBW	4	1.29 (0.66–2.53)	Random	0.46^#
Obese	3	2.84 (2.17–3.71)	Random	<0.00001
Overweight	3	1.38 (1.22–1.43)	Fixed	<0.00001
Adjustment for confounders Alcohol use
Yes
EBW	4	1.77 (1.52–2.07)	Random	<0.00001
Obese	3	2.51 (2.21–2.86)	Fixed	<0.00001
Overweight	3	1.32 (1.21–1.43)	Fixed	<0.00001
NO
EBW	14	1.58 (1.29–1.94)	Random	<0.00001
Obese	11	2.49 (2.12–2.93)	Random	<0.00001
Overweight	11	1.36 (1.12–1.64)	Random	0.002
Smoking
Yes
EBW	10	1.57 (1.27–1.93)	Random	<0.0001
Obese	7	2.57 (2.20–2.89)	Random	<0.00001
Overweight	7	1.41 (1.23–1.60)	Random	<0.00001
NO
EBW	8	2.69 (1.35–2.13)	Random	<0.00001
Obese	7	2.44 (1.73–3.44)	Random	<0.00001
Overweight	7	1.20 (0.95–1.53)	Random	0.13^#
Diabetes
Yes
EBW	4	2.09 (1.72–2.54)	Random	<0.00001
Obese	3	2.76 (1.94–3.93)	Random	<0.00001
Overweight	3	1.37 (1.11–1.71)	Fixed	0.004
NO
EBW	14	1.50 (1.25–1.79)	Random	<0.00001
Obese	11	2.49 (1.99–3.11)	Random	<0.00001
Overweight	11	1.30 (1.11–1.52)	Random	0.001
HT
Yes
EBW	7	1.48 (1.11–1.98)	Random	0.008
Obese	6	2.51 (2.19–2.88)	Random	<0.00001
Overweight	6	1.31 (1.22–1.41)	Fixed	<0.00001
NO
EBW	11	1.69 (1.43–2.00)	Random	<0.00001
Obese	9	2.64 (1.92–3.62)	Random	<0.00001
Overweight	9	1.39 (1.06–1.82)	Random	0.02
Age at menarche ¹
Yes
EBW	6	1.63 (1.01–2.64)	Random	0.05
Obese	4	3.18 (2.25–4.48)	Random	<0.00001
Overweight	4	1.54 (1.18–2.00)	Random	0.01
NO
EBW	12	1.58 (1.39–1.80)	Random	<0.00001
Obese	10	2.31 (1.84–2.90)	Random	<0.00001
Overweight	10	1.23 (1.14–1.65)	Random	0.006
Age at menopause ²
Yes
EBW	7	1.79 (1.56–2.06)	Random	<0.00001
Obese	5	2.64 (2.27–3.07)	Random	<0.00001
Overweight	5	1.39 (1.17–1.65)	Random	0.0002
NO
EBW	11	1.53 (1.17–2.00)	Random	0.002
Obese	9	2.50 (1.90–3.30)	Random	<0.00001
Overweight	9	1.27 (1.04–1.56)	Random	0.02
Parity ³
Yes
EBW	7	1.62 (1.07–2.44)	Random	0.02
Obese	5	2.97 (2.26–3.92)	Random	<0.00001
Overweight	5	1.44 (1.14–1.82)	Random	0.003
NO
EBW	11	1.56 (1.37–1.78)	Random	<0.00001
Obese	9	2.31 (1.18–2.96)	Random	<0.00001
Overweight	9	1.25 (1.06–1.48)	Random	0.007
Age at first full term pregnancy ⁴
Yes
EBW	3	1.70 (1.29–2.25)	Random	0.00002
Obese	2	3.52 (1.45–8.52)	Random	0.005
Overweight	2	1.41 (1.17–1.66)	Random	0.0002
NO
EBW	15	1.60 (1.33–1.93)	Random	<0.00001
Obese	12	2.43 (1.98–2.98)	Random	<0.00001
Overweight	12	1.31 (1.13–1.51)	Random	0.0004

no statistically significant correlation.

13 year was used as a cut-off point.

49 year was used as a cut-off point.

0 was used as a cut-off point.

20 year was used as a cut-off point.

Fig. 2

Forest plot for the association between EBW and endometrial cancer risk. Risk ratios derive from the random effects model by Revman 5.0.

Fig. 3

Forest plot for the association between obesity and endometrial cancer risk. Risk ratios derive from the random effects model by Revman 5.0.

Fig. 4

Forest plot for the association between overweight subjects and endometrial cancer risk. Risk ratios derive from the random effects model by Revman 5.0.

Subgroup Meta-Analyses

We then conducted subgroup meta-analyses by study design, geographic locations, BMI ascertainment, and possible confounders (e.g. alcohol use, smoking status, history of diabetes, hormone therapy [HT], age at menarche (≥13 year cf <13 year) (15), age at menopause (≥49 year cf <49 year) (13) (the study have shown the mean age at menopause in this population to be about 49–50 years, thus, age 49 was used as a cut-off point in subgroup analyses based on age in women), parity (>0 cf 0) (15) and age at first full term pregnancy) <20 year cf ≥20 year) (24)).

Analyses by Study Design

When stratifying for study design, we found that there was a statistically significant association between the conditions of EBW, overweight or obesity and endometrial cancer risk in studies with a cohort design (for EBW RR=1.62, 95% CI, 1.14–2.30; for obesity RR=3.11, 95% CI, 2.62–3.69; for overweight RR=1.47, 95% CI, 1.22–1.78) as well as in studies with a case-control design (for EBW RR=1.60, 95% CI, 1.39–1.84; for obesity RR=2.06, 95% CI, 1.61–2.63; for overweight RR=1.18, 95% CI, 0.96–1.45; for this last correlation there was no statistical significance).

Analyses by Geographic Locations and BMI Ascertainment

When stratifying for geographic locations, there was a statistically significant correlation between the conditions of EBW, overweight or obesity and endometrial cancer risk for Asians (for EBW RR=1.68, 95% CI, 1.35–2.09; for obesity RR=2.65, 95% CI, 1.34–5.24; for overweight RR=1.80, 95% CI, 0.89–3.67) and non-Asians (for EBW RR=1.59, 95% CI, 1.29–1.95; for obese, RR=2.52, 95% CI, 2.04–3.12; for overweight RR=1.29, 95% CI, 1.12–1.48). When stratifying for BMI ascertainment, there was a statistically significant correlation between the conditions of EBW, overweight or obesity and endometrial cancer risk for self-reported BMI (for EBW RR=1.70, 95% CI, 1.49–1.94; for obesity RR=2.46, 95% CI, 1.96–3.08; for overweight RR=1.29, 95% CI, 1.10–1.52) and for measured BMI (for EBW RR=1.29, 95% CI, 0.66–2.53; for obesity RR=2.84, 95% CI, 2.17–3.71; for overweight RR=1.38, 95% CI, 1.22–1.43).

Analyses by Confounders

When stratifying for the adjustment for confounders, there was a statistically significant correlation between EBW and endometrial cancer risk for a number of factors: alcohol use (for EBW RR=1.77, 95% CI, 1.52–2.07) or no alcohol use (for EBW RR=1.58, 95% CI, 1.29–1.94); smokers (for EBW RR=1.57, 95% CI, 1.27–1.93) or non-smokers (for EBW RR=2.69, 95% CI, 1.35–2.13); history of diabetes (for EBW RR=2.09, 95% CI, 1.72–2.54) or no history of diabetes (for EBW RR=1.50, 95% CI, 1.25–1.79); HT (for EBW RR=1.48, 95% CI, 1.11–1.98) or no HT (for EBW RR=1.69, 95% CI, 1.43–2.00), age at menarche (13 year was used as a cut-off point) (for EBW, RR=1.63, 95% CI, 1.01–2.64) or no age at menarche (for EBW RR=1.58, 95% CI, 1.39–1.80); age at menopause (49 year was used as a cut-off point) (for EBW RR=1.79, 95% CI, 1.56–2.06) or no age at menopause (for EBW RR=1.53, 95% CI, 1.17–2.00); parity (0 was used as a cut-off point) (for EBW RR=1.62, 95% CI, 1.07–2.44) or no parity (for EBW RR=1.56, 95% CI, 1.37–1.78); and, finally, age at first full term pregnancy (20 year was used as a cut-off point) (for EBW RR=1.70, 95% CI, 1.29–2.25) or no age at first full term pregnancy (for EBW RR=1.60, 95% CI, 1.33–1.93).

Test for Heterogeneity

Statistically significant heterogeneity was observed among the included studies (for EBW, x²=207.82, p<0.00001, I²=92%; for obesity, x²=141.15, p<0.00001, I²=91%; for overweight, x²=48.18, p<0.00001, I²=73%). In addition, we found evidence of an existing publication bias, as shown by the funnel plot in Figure 5.

Fig 5

Funnel plot for all studies included in this meta-analysis of EBW by endometrial cancer risk (Revman 5.0).

Discussion

Obesity is one of the most common disorders in medical practice and is a major public health problem. A recent national survey in the US found that approximately 50% of women are overweight and 25% are obese. Previous studies have shown that obesity is associated with an increased risk of various types of cancers, among which cancers of the stomach (29 –32), liver (33 –37), ovaries (38 –41), pancreas (42 –45), and colon (46 –49). Our analysis of published cohort and case-control studies also support a positive association between EBW, overweight or obesity and the risk of endometrial cancer. The strength of the association increased with increasing BMI values. Moreover, a previous review also confirmed such association with endometrial cancer (50).

In a subgroup analysis stratified on basis of the study design, we found a statistically significant link between EBW or obesity and the risk of endometrial cancer in cohort design and case-control design respectively. When stratifying the studies for either Asian or non-Asian populations, we found that the conditions of EBW, overweight and obesity were associated with a statistically significant increased risk of endometrial cancer among both Asians and non-Asians. Setiawan et al (51) showed that the interracial differences in endometrial cancer risk do not appear to be explained by differences in the distribution of known risk factors among women of different races/ethnicities.

This meta-analysis was based on data from prospective studies. Individual studies may have failed to control for potential confounders. Several studies adjusted their analysis for smoking status (11 –13, 15 –20, 27), alcohol consumption (12, 16, 19, 20), diabetes (13, 20, 22, 27), and/or HT (12, 13, 15, 16, 19, 20, 27), since these factors have been associated with the risk of endometrial cancer in epidemiological studies. We found that there was a statistically significant correlation between EBW and endometrial cancer risk for smoking subjects. Lindemann and colleagues (52) also reached a similar conclusion. We also found a statistically significant correlation between EBW and endometrial cancer risk for alcohol use. With regard to the history of diabetes, Saltzman et al (53) found that type 2 diabetes was associated with endometrial cancer in women with a BMI <35 but not in women with a BMI ≥35. A history of diabetes was associated with a twofold increased risk of endometrial cancer among hypertensive women, but no association was observed among non hypertensive women. Type 2 diabetes seems to be associated with endometrial cancer irrespective of the presence of other risk factors for this disease, except, possibly, hypertension and extreme obesity. We also found a statistically significant correlation between EBW and endometrial cancer risk for HT. In the European Prospective Investigation into Cancer and Nutrition cohort (EPIC), obesity was positively associated with endometrial cancer risk in postmenopausal women who never used HT, but was not associated with an increased risk in those who ever used HT (54). Age at menarche, age at menopause, parity, and age at first full term pregnancy are intrinsic confounders for the relationship between obesity and endometrial cancer. Therefore, ignoring these confounders could make the association between obesity and endometrial cancer less strong.

There are several potential mechanisms by which EBW may increase the risk of endometrial cancer. Weight gain is largely reflected in body fat accumulation. Excessive conversion of androstenedione to oestrone, and decreased serum levels of the sex hormone-binding globulin (SHBG), related to obesity, create an estrogen-rich environment that stimulates the development of endometrial cancer cells (55). Premenopausal obese women are likely to have anovulatory cycles and decreased levels of progesterone. In particular, obese women are exposed to prolonged unopposed estrogens during early adulthood and, as a result, might have an increased risk of endometrial cancer (56). In postmenopausal women, the mechanism of endometrial cancer development can be explained by the peripheral conversion of androgens into estrogens caused by an increase in peripheral fat stored. This hormonal alteration stimulates the proliferation of endometrial cells by inhibiting apoptosis, thus promoting angiogenesis (57). Other possible mechanisms may also involve: (i) insulin resistance and hyperinsulinemia (58); (ii) adiponectin (59); (iii) sex hormones, adipokines, inflammatory factors, and cytokines.

To our knowledge, the strengths of this study include: the consistency of a positive association between EBW and endometrial cancer across some high-quality studies with different patient populations; the report of an increasingly significant association with increasing BMI values; the definition of comparable BMI categories for each study; the ability to stratify by study design, geographic locations, and BMI ascertainment; and, finally, the ability to assess the effect of including only estimates adjusted for potential confounders.

In summary, this meta-analysis of prospective studies indicates that the conditions of EBW, overweight, or obesity are positively associated with the risk of endometrial cancer. Additionally, the strength of the association increased with increasing BMI. Nevertheless, more research is needed to unravel the underlying mechanisms of this association and to determine whether interventions aimed at lowering the body weight can decrease the risk of endometrial cancer.

Footnotes

Acknowledgments

The authors thank Juyi Li and Jilian Chen for the technical assistance and the whole medical department for the valuable discussions.

Financial Support: This work was supported by the Key Project of the Anhui Universities Provincial Natural Science Research (KJ2010A205), and the Annual Research Project of Anhui Science and Technology Agency (1207403074).

Conflict of Interest Statements: The authors have declared that no conflict of interest exists. The authors warrant that this contribution is original, that they have not granted or assigned any rights in the article to any other person or entity, that the article is copyrightable and that it does not infringe upon any copyright, trademark, or patent.

References

Sankaranarayanan

, Ferlay

Worldwide burden of gynaecological cancer: the size of the problem.

Best Pract Res Clin Obstet Gynaecol 2006; 20: 207–25.

Zullo

, Palomba

, Russo

A prospective randomized comparison between laparoscopic and laparotomic approaches in women with early stage endometrial cancer: a focus on the quality of life.

Am J Obstet Gynecol 2005; 193: 1344–52.

Jemal

, Siegel

, Ward

Cancer statistics 2006.

CA Cancer J Clin 2006; 56: 106–30.

Weir

, Thun

, Hankey

Annual report to the nation on the status of cancer, 1975–2000, featuring the uses of surveillance data for cancer prevention and control.

J Natl Cancer Inst 2003; 95: 1276–99.

Di Saia

, Brewster

WR.

Hormone replacement therapy for survivors of breast and endometrial cancer.

Curr Oncol Rep 2002; 4: 152–8.

Hedley

, Ogden

, Johnson

Prevalence of overweight and obesity among US children, adolescents, and adults, 1999–2002.

JAMA 2004; 291: 2847–50.

McCourt

, Mutch

, Gibb

Body mass index: relationship to clinical, pathologic and features of microsatellite instability in endometrial cancer.

Gynecol Oncol 2007; 104: 535–9.

Schottenfeld

Epidemiology of endometrial neoplasia.

J Cell Biochem Suppl 1995; 23: 151–9.

Okamura

, Tsubono

, Kiyoshi

Lactation and risk of endometrial cancer in Japan: a case-control study.

Tohoku J Exp Med 2006; 208: 109–15.

10.

Higgins

, Thompson

SG.

Quantifying heterogeneity in a meta- analysis.

Stat Med 2002; 21: 1539–58.

11.

Lukanova

, Bjor

, Kaaks

Body mass index and cancer: Results from the Northern Sweden Health and Disease Cohort.

Int J Cancer 2006; 118: 458–66.

12.

Reeves

, Pirie

, Beral

Cancer incidence and mortality in relation to body mass index in the Million Women Study: cohort study.

BMJ 2007; 335: 1134–44.

13.

Chang

, Lacey

Jr , Brinton

Lifetime weight history and endometrial cancer risk by health study and type of menopausal hormone use in the NIH-AARP diet.

Cancer Epidemiol Biomarkers Prev 2007; 16: 723–30.

14.

Canchola

, Chang

, Bernstein

Body size and the risk of endometrial cancer by hormone therapy use in postmenopausal women in the California Teachers Study cohort.

Cancer Causes Control 2010; 21:1407–16.

15.

Jain

, Rohan

, Howe

A cohort study of nutritional factors and endometrial cancer.

Eur J Epidemiol 2000; 16: 899–905.

16.

Conroy

, Sattelmair

, Cook

Physical activity, adiposity, and risk of endometrial cancer.

Cancer Causes Control 2009; 20: 1107–15.

17.

Schouten

, Goldbohm

, van den Brandt

PA.

Anthropometry, physical activity, and endometrial cancer risk: results from the Netherlands cohort study.

J Natl Cancer Inst 2004; 96: 1635–8.

18.

, Xiang

, Zheng

Weight history and risk of endometrial cancer among Chinese women.

Int J Epidemiol 2006; 35: 159–66.

19.

, Risch

, Irwin

Long-term overweight and weight gain in early adulthood in association with risk of endometrial cancer.

Int J Cancer 2011; 129 (5): 1237–43.

20.

Hosono

, Matsuo

, Hirose

Weight gain during adulthood and body weight at age 20 are associated with the risk of endometrial cancer in Japanese women.

J Epidemiol 2011; 21: 466–73.

21.

Folsom

, Kaye

, Potter

Association of incident carcinoma of the endometrium with body weight and fat distribution in older women: early findings of the Iowa Women's Health Study.

Cancer Res 1989; 49: 6828–31.

22.

Charneco

, Ortiz

, Venegas-Ríos

Clinic-based case-control study of the association between body mass index and endometrial cancer in Puerto Rican women.

P R Health Sci J 2010; 29: 272–8.

23.

Dal Maso

, Augustin

, Karalis

Circulating adiponectin and endometrial cancer risk.

J Clin Endocrinol Metab 2004; 89: 1160–3.

24.

John

, Koo

, Horn-Ross

PL.

Lifetime physical activity and risk of endometrial cancer.

Cancer Epidemiol Biomarkers Prev 2010; 19: 1276–83.

25.

Fortuny

, Sima

, Bayuga

Risk of endometrial cancer in relation to medical conditions and medication use.

Cancer Epidemiol Biomarkers Prev 2009; 18: 1448–56.

26.

Trentham-Dietz

, Nichols

, Hampton

Weight change and risk of endometrial cancer.

Int J Epidemiol 2006; 35: 151–8.

27.

Potischman

, Hoover

, Brinton

Case-control study of endogenous steroid hormones and endometrial cancer.

J Natl Cancer Inst 1996; 88: 1127–35.

28.

Abnet

, Freedman

, Hollenbeck

A prospective study of BMI and risk of oesophageal and gastric adenocarcinoma.

Eur J Cancer 2008; 44: 465–71.

29.

Persson

, Inoue

, Sasazuki

Female reproductive factors and the risk of gastric cancer in a large-scale population-based cohort study in Japan (JPHC study).

Eur J Cancer Prev 2008; 17: 345–53.

30.

Sjodahl

, Jia

, Vatten

Body mass and physical activity and risk of gastric cancer in a population-based cohort study in Norway.

Cancer Epidemiol Biomark Prev 2008; 17: 135–40.

31.

Tanaka

, Nagata

, Oba

Prospective cohort study of body mass index in a dolescence and death from stomach cancer in Japan.

Cancer Sci 2007; 98: 1785–9.

32.

Larsson

, Wolk

Overweight, obesity and risk of liver cancer: a meta-analysis of cohort studies.

Br J Cancer 2007; 97: 1005–8.

33.

Batty

, Barzi

, Huxley

Obesity and liver cancer mortality in Asia: the Asia Pacific Cohort Studies Collaboration.

Cancer Epidemiol 2009; 33: 469–72.

34.

Loomba

, Yang

, Su

Obesity and alcohol synergize to increase the risk of incident hepatocellular carcinoma in men.

Clin Gastroenterol Hepatol 2010; 8: 891–8.

35.

Borena

, Strohmaier

, Lukanova

Metabolic risk factors and primary liver cancer in a prospective study of 578,700 adults.

Int J Cancer 2012; 131: 193–200.

36.

Kant

, Hull

MA.

Excess body weight and obesity–the link with gastrointestinal and hepatobiliary cancer.

Nat Rev Gastroenterol Hepatol 2011; 8: 224–38.

37.

Olsen

, Green

, Whiteman

Obesity and the risk of epithelial ovarian cancer: a systematic review and meta-analysis.

Eur J Cancer 2007; 43: 690–709.

38.

Pavelka

, Brown

, Karlan

Effect of obesity on survival in epithelial ovarian cancer.

Cancer 2006; 107: 1520–4.

39.

Skirnisdottir

, Sorbe

Prognostic impact of body mass index and effect of overweight and obesity on surgical and adjuvant treatment in early-stage epithelial ovarian cancer.

Int J Gynecol Cancer 2008; 18: 345–51.

40.

Skirnisdottir

, Sorbe

Body mass index as a prognostic factor in epithelial ovarian cancer and correlation with clinico-pathological factors.

Acta Obstet Gynecol Scand 2010; 89: 101–7.

41.

Nothlings

, Wilkens

, Murphy

Body mass index and physical activity as risk factors for pancreatic cancer: the Multiethnic Cohort Study.

Cancer Causes Control 2007; 18: 165–75.

42.

Luo

, Iwasaki

, Inoue

Body mass index, physical activity and the risk of pancreatic cancer in relation to smoking status and history of diabetes: a large-scale population-based cohort study in Japan—the JPHC study.

Cancer Causes Control 2007; 18: 603–12.

43.

Luo

, Margolis

, Adami

Obesity and risk of pancreatic cancer among postmenopausal women: the Women's Health Initiative (United States).

Br J Cancer 2008; 99: 527–31.

44.

Johansen

, Borgstrom

, Lindkvist

Different markers of alcohol consumption, smoking and body mass index in relation to risk of pancreatic cancer.

A prospective cohort study with in the Malmo Preventive Project. Pancreatology 2009; 9: 677–86.

45.

Renehan

, Tyson

, Egger

Body mass index and incidence of cancer: a systematic review and meta-analysis of prospective observational studies.

Lancet 2008; 371: 569–78.

46.

Bowers

, Albanes

, Limburg

A prospective study of anthropometric and clinical measurements associated with insulin resistance syndrome and colorectal cancer in male smokers.

Am J Epidemiol 2006; 164: 652–64.

47.

MacInnis

, English

, Haydon

, Body size and composition and risk of rectal cancer (Australia). Cancer Causes Control 2006, 17: 1291–7.

48.

Thygesen

, Grønbaek

, Johansen

Prospective weight change and colon cancer risk in male US health professionals.

Int J Cancer 2008; 123: 1160–5.

49.

Setiawan

, Pike

, Kolonel

Racial/ethnic differences in endometrial cancer risk: the multiethnic cohort study.

Am J Epidemiol 2007; 165: 262–70.

50.

Lindemann

, Vatten

, Ellstrom-Engh

Body mass, diabetes and smoking, and endometrial cancer risk: a follow-up study.

Br J Cancer 2008; 98: 1582–5.

51.

Saltzman

, Doherty

, Hill

Diabetes and endometrial cancer: an evaluation of the modifying effects of other known risk factors.

Am J Epidemiol 2008; 167: 607–14.

52.

Friedenreich

, Cust

, Lahmann

Anthropometric factors and risk of endometrial cancer: the European prospective investigation into cancer and nutrition.

Cancer Causes Control 2007; 18: 399–413.

53.

Evans

, Hoffmann

, Kalkhoff

Relationship of androgenic activity to body fat topography, fat cell morphology, and metabolic aberrations in premenopausal women.

J Clin Endocrinol Metab 1983; 57: 304–9.

54.

Fader

, Arriba

, Frasure

Endometrial cancer and obesity: epidemiology, biomarkers, Prevention and survivorship.

Gynecol Oncol 2009; 114: 121–7.

55.

Bakkum-Gamez

, González-Bosquet

, Laack

Current issues in the management of endometrial cancer.

Mayo Clin Proc 2008; 83: 97–112.

56.

Berstein

, Kvatchevskaya

, Poroshina

Insulin resistance, its consequences for the clinical course of the disease, and possibilities of correction in endometrial cancer.

J Cancer Res Clin Oncol 2004; 130: 687–93.

57.

Soliman

, Wu

, Tortolero-Luna

Association between adiponectin, insulin resistance, and endometrial cancer.

Cancer 2006; 106: 2376–81.